Objective estimation of subjective image quality assessment using multiparameter prediction

What is it about?

Objective evaluation of a subjective image quality assessment plays a significant role in the various image processing applications, such as compression, interpolation and noise reduction. The subjective image quality assessment does not only depend on some objective measurable artefacts, but also on image content and kind of distortions. Thus, a multi-parameter prediction of the objective image quality assessment is proposed in this paper. The prediction parameters are found minimizing the mean square error related to the known subjective image quality measure (DMOS). This approach includes mostly used image quality metrics (PSNR, MSSIM, FSIM, VQM) and two-dimensional image quality metrics (2D IQM). The proposed multi-parameter prediction has been verified on the test image database (LIVE) for compression, noise and blur distortions with available subjective image quality measures (DMOS). More reliable estimations are obtained using multi-parameter prediction instead of only one measure. The best results are reached when an image content indicator is combined with the 2D IQM measure separately for different kinds of distortions.

Featured Image

Photo by Joanna Nix on Unsplash

Photo by Joanna Nix on Unsplash

Why is it important?

Prediction based on objective measures shows non-linear dependence. Satisfactory results are reached by square polynomial approximation. Better predictions are reached when the kind of distortions has been considered. However, separate predictions require different sets of prediction coefficients for each kind of distortions; significant improvements of predictions are achieved including the image content indicator. As the quantifier for image content the percent of pixels in edge areas is used. • in almost all cases the prediction based on 2D measures provides the most reliable results.

Perspectives